Fig 1.
Block diagram of the proposed human anxiety classification scheme in response to exposure therapy.
Fig 2.
Timing diagram of the data acquisition protocol followed in DASPS database.
Fig 3.
Raw EEG data obtained from the 14 channels of the Emotiv EPOC headset.
Fig 4.
Percentage of instances using SAM score based labeling for (a) Two-level (b) Four-level and HAM-A based lableling for (c) Two-level (d) Four-level anxiety classification.
Fig 5.
Box-and-whisker diagram of the power obtained from each channel of the Emotiv EPOC headset for binary anxiety classification.
Fig 6.
Box-and-whisker diagram of the power obtained from each channel of the Emotiv EPOC headset for four level anxiety classification.
Fig 7.
Classification accuracy of DT, k-NN, SVM, MLP and RF classifier using features from different frequency band combinations of EEG signal using SAM score based labeling for (a) Two-levels (b) Four-levels anxiety classification.
Fig 8.
Classification accuracy of DT, k-NN, SVM, MLP and RF classifier using features from different frequency band combinations of EEG signal using HAM-A based labeling for (a) Two-levels (b) Four-levels anxiety classification.
Table 1.
Performance comparison of proposed anxiety classification framework using both SAM and HAM-A based labeling for two and four levels in terms of different machine learning classifiers, Feature Vector Length (FVL), accuracy, F-value, and Kappa values (K).
Table 2.
Confusion matrix for two level anxiety classification for DT, kNN, SVM, MLP and RF classifier using SAM based labeling.
Table 3.
Confusion matrix for two level anxiety classification for DT, kNN, SVM, MLP and RF classifier using HAM-A based labeling.
Table 4.
Confusion matrix for four level anxiety classification for DT, k-NN, SVM, MLP and RF classifier using SAM based labeling.
Table 5.
Confusion matrix for four level anxiety classification for DT, k-NN, SVM, MLP and RF classifier using HAM-A based labeling.
Table 6.
Performance comparison of the proposed framework with the state-of-the-art methods for human state anxiety classification.